Project Summary The drug-resistant form of tuberculosis (TB) has the potential to reduce quality of life and increase the likelihood of mortality. Multidrug-resistant TB (MDR-TB) is treated with at least 6 months of injectable aminoglycoside (AG). Among MDR-TB patients, permanent hearing loss due to the ototoxic effects from long-term use of AGs is profound and may lead to early AG discontinuation. Despite the known risk of AG-induced hearing loss, selections and availability of less ototoxic antibiotics or monitoring of hearing loss are constrained because of limited public health resources in South Africa. It is known that excessive AG concentration, renal impairment, coinfection with HIV with a certain class of antiretroviral therapy (ART), severe systemic inflammation, malnutrition, preexisting hearing loss, and several demographic factors worsen a patient?s potential risk for AG- induced hearing loss during MDR-TB treatment. Among these risk factors, AG concentration needs to receive more attention because ototoxicity is dose-dependent. Maintaining therapeutic AG concentration aids in hearing loss prevention and cure of MDR-TB, but daily blood testing for drug concentration is impractical in South Africa due to financial constraints. Therefore, this proposed study is aiming to estimate the risk of AG-induced hearing loss among MDR-TB infected individuals in South Africa. This study will (1) develop a prediction model of developing AG-induced hearing loss from MDR-TB treatment, and (2) explore the prognostic impact of cumulative AG exposure on AG-induced hearing loss in MDR-TB patients after injectable AG therapy initiation. Cumulative AG exposure will be utilized as a surrogate measure of AG concentration and of the tested time- varying effect of cumulative AG exposure on the risk of AG-induced hearing loss to be introduced into the prediction model. The proposed study is a secondary analysis of an ongoing cluster randomized clinical trial (R01 AI104488) to improve treatment outcomes in individuals with MDR-TB infection in South Africa. All subjects enrolled in the parent study who are self-identified black South African receiving intramuscular kanamycin or amikacin will be included for the proposed study. All key patient details will be abstracted from the parent study including basic demographics with MDR-TB/HIV history and clinical parameters, such as hearing, ART, CD4, estimated glomerular filtration rate (eGFR), lung cavities, BMI, serum albumin, as well as MDR-TB regimen including dosing and frequency. Finally, the developed prediction model will assist decision makers in prioritizing vulnerability to hearing loss and providing less ototoxic drugs for prevention of hearing loss within limited resources. The findings will contribute to reducing AG-induced hearing loss by developing a prediction model and filling the gaps in personalized interventions to prevent hearing disability in underserved populations.